Re: [PATCH v2] mm/page_alloc: bail out on fatal signal during reclaim/compaction retry attempt

On Wed 2021-05-19 20:32 +0100, Matthew Wilcox wrote:

On Wed, May 19, 2021 at 08:23:21PM +0100, Aaron Tomlin wrote:

quoted

+++ b/mm/page_alloc.c

@@ -4252,6 +4252,9 @@ should_compact_retry(struct alloc_context *ac, int order, int alloc_flags,
 	if (!order)
 		return false;
 
+	if (fatal_signal_pending(current))
+		goto out;

I think 'goto out' will be confusing.  It'll output a tracepoint, which
isn't going to record that a fatal signal is pending, so it'll cause
some head scratching for someone looking through the traces.  I
think we should just return false here and skip the tracepoint.

I agree. Having said this, I do plan to send a patch to illustrate why
compaction should not retry to avoid possible confusion. Nevertheless, I am
happy to modify as per your request.


Kind regards,

-- 
Aaron Tomlin